From flat sheets to complex structures

One major way that developing embryos build their organs is through furrowing — that is, they form pockets in tissues, which eventually become the sites of folds. "Just as a flat sheet of paper can be folded into a crane, a flat embryonic tissue can be folded into the precursor of an organ," said Andrew Countryman, a doctoral student in biomedical engineering at Columbia and the study's first author.

Previous research has developed many tools for manipulating the proteins and other molecules that direct how cells behave. However, scientists lacked similar techniques for systematically controlling the mechanical forces that ultimately shape embryos. 

In the new study, Kasza, Countryman, and their colleagues experimented with the fruit fly, a common lab animal. "As developmental processes and machinery are highly conserved across animals, these findings in fruit flies provide insight into development in all animals, including humans," Countryman said.

Light-sensitive tools built with CRISPR

The researchers tinkered with proteins that cells use to generate mechanical forces, making these molecules responsive to light. By shining patterns of specific wavelengths of light on fruit fly embryos genetically modified to produce these proteins, they could in turn control patterns of forces during their development. 

The new study used the gene-editing system CRISPR-Cas9 to add a light-sensitive module to genes that naturally exist in fruit flies. The resulting molecules are the first tools that let scientists use light to control an animal's own genes to direct mechanical forces in live embryos. They are also the first tools that enable scientists to employ light to control cell-generated forces in a tunable way, instead of just switching such forces on and off, Countryman said.

The researchers specifically modified proteins that help cells contract, one method by which tissues can generate furrows. The resulting tools, called endogenous OptoRhoGEFs, helped the scientists discover that the depth of a furrow depends on the amount of these contraction-linked proteins that get summoned to a cell's membrane. They also found that stiff layers of proteins within embryos could dramatically influence the ways in which tissues furrowed.

Implications for human health

"Similarly to fly embryos, human embryos extensively employ furrowing processes during development," Countryman said. "A failure of tissues to furrow properly is associated with common and devastating congenital disorders, such as spina bifida. Improved understanding of developmental processes will help identify and treat these conditions."

This new technique may one day help scientists better analyze tissue and organ development and disease, using light to fold basic sheets of cells into complex 3D structures in the lab instead of the more complex environments inside living animals, Countryman said.

In addition, "small, controllable, cell-based machines have promising use in medical contexts, where they can serve as biocompatible probes during medical procedures," he added. "They could also be used as small, aqueous, remotely pilotable vehicles to explore and survey new environments."

In the future, the researchers hope to use their new strategy to examine other ways in which tissues furrow, as well as tissue behaviors other than furrowing, such as bending, stretching, and flowing. "These basic modes of tissue deformation are used in different combinations and sequences to build a wide variety of tissues, organs, and body forms," Countryman said.

Lead Photo Caption: Re-engineering force-regulating proteins inside cells to control their behavior with specific wavelengths of light. Side view of a small group of optogenetically activated cells forming a furrow toward the inside of the embryo (top image, far left); Side view of a large group of optogenetically activated cells bending toward the outside of the embryo (bottom image, far left); Top view of a large group of optogenetically activated cells bending toward the outside of the embryo (center); Patterns of myosin–a contractile protein–associated with optogenetic activation of a large group of cells (far right). 

Lead Photo Credit: Andrew Countryman/Kasza lab

About The Study

Journal: Nature Communications

Title: Endogenous OptoRhoGEFs reveal biophysical principles of epithelial tissue furrowing

DOI: 10.1038/s41467-025-62483-6

Authors: Andrew D. Countryman, Caroline A. Doherty, R. Marisol Herrera-Perez, Karen E. Kasza.

Funding/Acknowledgements: The researchers thank Stas Shvartsman and Liz Gavis for their contributions to conceptualization of the endogenous optogenetic tools and for helpful discussions. They thank Bex Pendrak, Sameer Thukral, and Kasza Lab members for helpful discussions. Stocks obtained from the Bloomington Drosophila Stock Center (NIH P40OD018537) were used in this study. This work was performed in part at the Live Imaging and Bioenergetics Facility at the Advanced Science Research Center at The Graduate Center of the City University of New York. This work was supported by NIH Grant R35GM138380 to Karen E. Kasza and NIH Grant 1F31HD118793-01 to Andrew D. Countryman. Karen E. Kasza holds an NSF CAREER Award, Packard Fellowship, and Sloan Research Fellowship in Physics.

The authors declare no competing interests.

This sort of work is (pardon the pun) in its infancy. While digital twins of some organs, like the heart, are already advanced, we’re still doing the fundamental research to understand the uterus, cervix, and overall functioning of the women’s reproductive system. 

It’s a big deal because these twins are only as good as the underlying data. Unlike in aerospace engineering, there’s no detailed atlas of the orientation of fibers that comprise the cervix. We don’t yet have standardized measurements of how the uterus stretches week by week across pregnancy. Shockingly, information about fundamental properties of pregnancy varies widely in the literature, if they’ve been documented at all.

That lack of information is a problem for researchers, physicians, industry, and patients. Without foundational knowledge, femtech startups are left taking guesses or burning time and capital answering basic questions. Founders regularly ask me questions about the size and shape of the uterus and cervix or how they’re packaged in the vaginal canal. These inventors are building important technologies like more effective contraception devices and treatments to prevent preterm birth. It’s vital for them to have access to high-quality information.

Image

To support safe, effective innovation in femtech and better clinical outcomes, we need sustained investment in the full research pipeline: from curiosity-driven science to technology transfer and clinical applications. That means support from philanthropy and industry as well as from the federal government. One reason my lab has been able to continue this work over the years is the ongoing support of the Iris Fund, a foundation focused on advancing research in preterm birth and supporting families that experience high-risk pregnancies. When other sources of funding have fallen short, the Iris Fund has stepped in to fill in the gaps. 

We’ve figured out how to model jet wings and SUVs — understanding how the uterus stretches through the course of pregnancy is within our grasp. But we can’t skip straight to product design. First, we need the science.

Kristin Myers is a professor of mechanical engineering, director of the Myers Soft Tissue Lab at Columbia University, and an expert in women's health engineering research.

Lead Photo Caption: Kristin Myers spoke on a panel about digital twins in medicine at the 2025 Aspen Ideas Health festival.

Lead Photo Credit: Aspen Institute

The researchers showed how individual Truss Links self-assembled into two-dimensional shapes that then could morph into three-dimensional robots. These robots then further improved themselves by integrating new parts, effectively "growing" into more capable machines. For example, a 3D tetrahedron shaped robot integrated an additional link that it could use like a walking stick to increase its downhill speed by more than 66.5%. 

"Robot minds have moved forward by leaps and bounds in the past decade through machine learning, but robot bodies are still monolithic, unadaptive, and unrecyclable,” says Hod Lipson, co-author and James and Sally Scapa Professor of Innovation and chair of the Department of Mechanical Engineering at Columbia University, and director of the Creative Machines lab where the work was done. “Biological bodies, in contrast, are all about adaptation - lifeforms, can grow, heal, and adapt. In large part, this ability stems from the modular nature of biology that can use and reuse modules (amino acids) from other lifeforms. Ultimately, we’ll have to get robots to do the same - to learn to use and reuse parts from other robots. You can think of this nascent field as a form of ‘machine metabolism.’”

Researchers envision future robot ecologies where machines independently maintain themselves, growing and adapting to unforeseen tasks and environments. By imitating nature's approach—building complex structures from simple building blocks—robot metabolism paves the way for autonomous robots capable of physical development and long-term resilience.

"Robot Metabolism provides a digital interface to the physical world and allows AI to not only advance cognitively, but physically—creating an entirely new dimension of autonomy," says Wyder. "Initially, systems capable of Robot Metabolism will be used in specialized applications such as disaster recovery or space exploration. Ultimately, it opens up the potential for a world where AI can build physical structures or robots just as it today writes or rearranges the words in your email."

Lipson concludes with caution: “The image of self-reproducing robots conjures some bad sci-fi scenarios. But the reality is that as we hand off more and more of our lives to robots - from driverless cars to automated manufacturing, and even defense and space exploration. Who is going to take care of these robots? We can’t rely on humans to maintain these machines. Robots must ultimately learn to take care of themselves.”

This work was supported by the NSF AI Institute in Dynamic Systems (NSF NIAIR COLUM 5260404, SPONS GG017178), NSF NRI (NSF NRI COLUM 5216104, SPONS GG015647), and DARPA TRADES (DARPA TRADES COLUM 5216104, SPONS GG012620).

Lead Photo Caption: These Truss Links self-assembled to form a tetrahedron.
Lead Photo Credit: Creative Machines Lab

About the Study

Title: “Robot Metabolism: Towards machines that can grow by consuming other machines”

Authors: Philippe Martin Wyder, Riyaan Bakhda, Meiqi Zhao, Quinn A. Booth, Matthew E. Modi, Andrew Song, Simon Kang, Jiahao Wu, Priya Patel, Robert T. Kasumi, David Yi, Nihar Niraj Garg, Pranav Jhunjhunwala, Siddharth Bhutoria, Evan H. Tong, Yuhang Hu, Judah Goldfeder, Omer Mustel, Donghan Kim, and Hod Lipson

For interview/questions contact: Philippe Wyder: [email protected] +1 (347) 604 4450

Hod Lipson: [email protected] +1 (607) 592 4383 

COI: The authors declare no financial or other conflicts of interest.

The successful launch capped off a yearslong effort by nearly 60 undergraduates to design, build, and test the rocket. The project came to fruition at the FAR-OUT 2024-2025 competition, where the rocket reached its target altitude for the first time in team history and achieved full recovery. The team shared video clips of the launch with Columbia Engineering faculty and supporters.

“This is an incredible achievement,” said Hod Lipson, James and Sally Scapa Professor of Innovation and chair of the Department of Mechanical Engineering. “It shows how our engineering education has accelerated from textbooks and lectures to record-breaking experiential learning. This is literally the next generation of rocket scientists.”

“It made my day to see our student team launch this newly designed rocket,” said Shih-Fu Chang, Dean of Columbia Engineering.

Now in its third consecutive year of successful launches, CSI Rockets has become a proving ground for student-led aerospace work. With this milestone behind them, the team is already planning its next move.

“We’re excited for what the next generation will accomplish,” said Sheehan. “This is just the beginning.”

Image

Getting to the Launchpad

The CSI Rockets team is composed entirely of undergraduate students, drawing from a wide range of disciplines and schools across Columbia University. While the group is rooted in engineering, it includes students from Columbia College, Columbia Engineering, and Barnard.

While most teams at FAR-OUT work on the same rocket for two or three years, the Columbia team builds a new rocket each year. 

“We do something different with the rocket every year,” said CSI Rockets co-lead Valentina Fichera, who graduated in 2025 with a degree in mechanical engineering. “Everyone was on board with transitioning to liquid oxygen — it forced us to push ourselves.”

This year’s team included about 60 members, with 19 traveling to the desert for the launch. Those undergraduate students were spread across specialized teams focused on propulsion, electronics, airframe, payload, and the combustion chamber. 

Image

Inside the Rocket

CSI Rockets had previously relied on nitrous oxide hybrid systems. But this year, the team switched to a liquid oxygen hybrid, which is a far more technically demanding system typically used in larger-scale commercial rockets. Few student teams even attempt liquid oxygen hybrids.

“We all wanted that experience,” Fichera said. 

“Liquid oxygen exists in a cryogenic state,” explained Sheehan. “So everything from storage to pressurization becomes more complex. We had to overhaul our entire propulsion system.” The team used gaseous nitrogen to pressurize the liquid oxygen and built new components to manage the system safely and effectively.

“We designed for a target apogee of 6,500 feet and reached nearly 5,700. It was our most accurate launch yet,” Fichera said. The launch itself came on the second-to-last window on the final day of the event, after multiple early-morning attempts. 

Fichera and Sheehan, who have interned at leading aerospace companies like Firefly and SpaceX, are pursuing engineering positions in industry.

“We spent a lot of time working on the rocket this year,” Fichera said with a laugh. 

CSI team members who attended the FAR-OUT rocketry competition:

• Michael Sheehan ’26SEAS (Team Co-Lead)
• Skylar Bogdanowitsch ’26SEAS (Team Co-Lead)
• Valentina Marini Fichera ’25SEAS (Team Co-Lead)
• Jorge Casas '23CC (Team Co-Lead)
• Aadam Awad ’26SEAS (Fluids Co-Lead)
• Vayu Singhal ’26SEAS (Fluids Co-Lead)
• Isabella Singleton ’27SEAS (Airframe Lead)
• Theo Lack ’26SEAS (Rising Fluids Co-Lead)
• Angela De Labra ’27SEAS (Rising Airframe Co-Lead)
• Alex Chen ’25SEAS (Electronics Co-Lead)
• Ania Krzyżańska ’25SEAS (Electronics Co-Lead)
• Tieqiong Zhang ’25SEAS (Electronics Co-Lead)
• Aruzhan Abil ’28CC (Rising Electronics Co-Lead)
• Joss Clegg ’28SEAS (Rising Electronics Co-Lead)
• Christopher Acosta ’25SEAS (Propulsion Combustion Chamber Co-Lead)
• Tingmeng Wang ’27SEAS (Propulsion Combustion Chamber Co-Lead)
• Raisa Effress ’27Barnard (Propulsion Combustion Chamber Co-Lead)
• Naomi Dreicer Liberman ’27SEAS (Rising Propulsion Combustion Chamber Co-Lead)
• Maria Cuevas ’26CC (Payload Co-Lead)
• Siroun Johnson ’26SEAS (Payload Co-Lead)

Lead Photo Credit: Courtesy of Columbia Space Initiative

Fresh from their win at the Millard Chan Tech Challenge, the team behind MilkShaker demonstrated a practical and science-backed solution to prevent mastitis, providing professional level care without sacrificing time or requiring special training. Another group of seniors in the Department of Mechanical Engineering showcased a prototype for the CareCruiser, a wheelchair-to-stroller attachment designed to give wheelchair-using parents and caregivers greater independence when navigating with strollers and carseats.

Several projects also tackled infrastructure design, including a civil engineering and engineering mechanics team that unveiled Rethinking Resilience. The project envisions a new sustainable, flood-resilient train station to replace the existing Ardsley-on-Hudson station along the MTA’s Metro-North Hudson Line. 

And in the gaming arena, a team from the Department of Electrical Engineering demonstrated their physical tank game using modified remote-controlled cars. By reconfiguring the electronics of remote-controlled cars, the team enabled Bluetooth-based control of both the motors and onboard lasers. Players score by hitting opposing cars with laser pulses, while any car that gets hit is temporarily disabled with a five-second freeze penalty.

Dean Shih-Fu Chang greeted the students and commended them on their year-long efforts. “You all have started with an idea and brought it to life. That’s what engineers do! We bring ideas to life . . . Today’s showcase truly represents our Engineering for Humanity vision. This is a milestone in your Columbia experience and one I hope you remember for years to come.”

Each year, Columbia Engineering seniors work on a capstone project, aka Senior Design. They are challenged to solve real-world problems with innovative solutions, rooted in their foundational math, science and engineering courses. 

Here’s a look at some of the students and their innovations at Senior Design Expo 2025. 

Masthead Caption: A student tries on LenScribe, a compact set of 3D-printed glasses designed for audio input, speech processing, and visual display

Masthead Credit: Timothy Lee/Columbia Engineering